Erik D. Goodman

Dimensionality reduction using genetic algorithms

Pattern recognition generally requires that objects be described in terms of a set of measurable features. The selection and quality of the features representing each pattern affect the success of subsequent classification. Feature extraction is the process of deriving new features from original features to reduce the cost of feature measurement, increase classifier efficiency, and allow higher accuracy. Many feature extraction techniques involve linear transformations of the original pattern vectors to new vectors of lower dimensionality. While this is useful for data visualization and classification efficiency, it does not necessarily reduce the number of features to be measured since each new feature may be a linear combination of all of the features in the original pattern vector. Here, we present a new approach to feature extraction in which feature selection and extraction and classifier training are performed simultaneously using a genetic algorithm. The genetic algorithm optimizes a feature weight vector used to scale the individual features in the original pattern vectors. A masking vector is also employed for simultaneous selection of a feature subset. We employ this technique in combination with the k nearest neighbor classification rule, and compare the results with classical feature selection and extraction techniques, including sequential floating forward feature selection, and linear discriminant analysis. We also present results for the identification of favorable water-binding sites on protein surfaces.

Coarse-grain parallel genetic algorithms: categorization and new approach

This paper describes a number of different coarse-grain GAs, including various migration strategies and connectivity schemes to address the premature convergence problem. These approaches are evaluated on a graph partitioning problem. Our experiments showed, first, that the sequential GAs used are not as effective as parallel GAs for this graph partition problem. Second, for coarse-grain GAs, the results indicate that using a large number of nodes and exchanging individuals asynchronously among them is very effective. Third, GAs that exchange solutions based on population similarity instead of a fixed connection topology get better results without any degradation in speed. Finally, we propose a new coarse-grained GA architecture, the Injection Island GA (iiGA). The preliminary results of iiGAs show them to be a promising new approach to coarse-grain GAs.<<ETX>>

The hierarchical fair competition (HFC) model for parallel evolutionary algorithms

The HFC model for evolutionary computation is inspired by the stratified competition often seen in society and biology. Subpopulations are stratified by fitness. Individuals move from low-fitness subpopulations to higher-fitness subpopulations if and only if they exceed the fitness-based admission threshold of the receiving subpopulation, but not of a higher one. HFCs balanced exploration and exploitation, while avoiding premature convergence, is shown on a genetic programming example.

The Hierarchical Fair Competition (HFC) Framework for Sustainable Evolutionary Algorithms

Many current Evolutionary Algorithms (EAs) suffer from a tendency to converge prematurely or stagnate without progress for complex problems. This may be due to the loss of or failure to discover certain valuable genetic material or the loss of the capability to discover new genetic material before convergence has limited the algorithms ability to search widely. In this paper, the Hierarchical Fair Competition (HFC) model, including several variants, is proposed as a generic framework for sustainable evolutionary search by transforming the convergent nature of the current EA framework into a non-convergent search process. That is, the structure of HFC does not allow the convergence of the population to the vicinity of any set of optimal or locally optimal solutions. The sustainable search capability of HFC is achieved by ensuring a continuous supply and the incorporation of genetic material in a hierarchical manner, and by culturing and maintaining, but continually renewing, populations of individuals of intermediate fitness levels. HFC employs an assembly-line structure in which subpopulations are hierarchically organized into different fitness levels, reducing the selection pressure within each subpopulation while maintaining the global selection pressure to help ensure the exploitation of the good genetic material found. Three EAs based on the HFC principle are tested - two on the even-10-parity genetic programming benchmark problem and a real-world analog circuit synthesis problem, and another on the HIFF genetic algorithm (GA) benchmark problem. The significant gain in robustness, scalability and efficiency by HFC, with little additional computing effort, and its tolerance of small population sizes, demonstrates its effectiveness on these problems and shows promise of its potential for improving other existing EAs for difficult problems. A paradigm shift from that of most EAs is proposed: rather than trying to escape from local optima or delay convergence at a local optimum, HFC allows the emergence of new optima continually in a bottom-up manner, maintaining low local selection pressure at all fitness levels, while fostering exploitation of high-fitness individuals through promotion to higher levels.

Swarmed feature selection

Feature selection is an important part of pattern recognition, helping to overcome the curse of dimensionality problem with classifiers, among other systems. In this work, we introduce a feature selection method using particle swarm optimization. Experiments using data of others and hyperspectral remote sensed data are used to measure the performance of the algorithm. Its comparison with a genetic algorithm is also shown.

Direct dimensional NC verification

Abstract A technique for automatic verification of three-axis numerically controlled (NC) milling programs is presented. Other methods proposed for NC verification rely on application of solid modelling technology. However, direct techniques suffer from heavy computational cost; and view-based techniques, while very efficient for simulation, are not capable of accurate dimensional NC verification. This article presents a unique approach to the problem based on direct comparison of the NC tool path program with a model of the desired part. An algorithm is presented that provides complete dimensional NC verification at a computational cost significantly less than direct solid modelling approaches. A software implementation of this algorithm produces graphical output depicting the desired part as shaded surfaces with out-of-tolerance areas highlighted. Several applications are presented that demonstrate verification of actual NC programs.

Toward a unified and automated design methodology for multi-domain dynamic systems using bond graphs and genetic programming

Abstract This paper suggests a unified and automated design methodology for synthesizing designs for multi-domain systems, such as mechatronic systems. A multi-domain dynamic system includes a mixture of electrical, mechanical, hydraulic, pneumatic, and/or thermal components, making it difficult use a single design tool to design a system to meet specified performance goals. The multi-domain design approach is not only efficient for mixed-domain problems, but is also useful for addressing separate single-domain design problems with a single tool. Bond graphs (BGs) are domain independent, allow free composition, and are efficient for classification and analysis of models, allowing rapid determination of various types of acceptability or feasibility of candidate designs. This can sharply reduce the time needed for analysis of designs that are infeasible or otherwise unattractive. Genetic programming is well recognized as a powerful tool for open-ended search. The combination of these two powerful methods is therefore an appropriate target for a better system for synthesis of complex multi-domain systems. The approach described here will evolve new designs (represented as BGs) with ever-improving performance, in an iterative loop of synthesis, analysis, and feedback to the synthesis process. The suggested design methodology has been applied here to three design examples. The first is a domain-independent eigenvalue placement design problem that is tested for some sample target sets of eigenvalues. The second is in the electrical domain––design of analog filters to achieve specified performance over a given frequency range. The third is in the electromechanical domain––redesign of a printer drive system to obtain desirable steady-state position of a rotational load.

Investigating Parallel Genetic Algorithms on Job Shop Scheduling Problems

This paper describes a GA for job shop scheduling problems. Using the Giffler and Thompson algorithm, we created two new operators, THX crossover and mutation, which better transmit temporal relationships in the schedule. The approach produced excellent results on standard benchmark job shop scheduling problems. We further tested many models and scales of parallel GAs in the context of job shop scheduling problems. In our experiments, the hybrid model consisting of coarse-grain GAs connected in a fine-grain-GA-style topology performed best, appearing to integrate successfully the advantages of coarse-grain and fine-grain GAs.

Evaluation of Injection Island GA Performance on Flywheel Design Optimisation

This paper first describes optimal design of elastic flywheels using an Injection Island Genetic Algorithm (iiGA). An iiGA in combination with a finite element code is used to search for shape variations to optimize the Specific Energy Density of flywheels (SED is the rotational energy stored per unit mass). iiGA’s seek solutions simultaneously at different levels of refinement of the problem representation (and correspondingly different definitions of the fitness function) in separate subpopulations (islands). Solutions are sought first at low levels of refinement with an axisymmetric plane stress finite element code for high-speed exploration of the coarse design space. Next, individuals are injected into populations with a higher level of resolution that uses an axisymmetric three-dimensional finite element model to “ fine-tune” the flywheel designs. Solutions found for these various “coarse” fitness functions on various nodes are injected into nodes that evaluate the ultimate fitness to be optimized Allowing subpopulations to explore different regions of the fitness space simultaneously allows relatively robust and efficient exploration in problems for which fitness evaluations are costly. First the paper treats a greatly simplified case — one for which all two million possible solutions were enumerated, yielding a known global optimum. Then the success and speed of many methods, including several variations of an iiGA, in finding this known global optimum are compared. The iiGA methods always found the global optimum, and the other methods never did. Hybridizing the iiGA with a local search operator and a Threshold Accepting (TA) search at the end of each generation provided the fastest solutions, without sacrificing robustness. Finally, a problem with a large design space is presented and results are compared for a hybrid iiGA to a parallel GA that uses a topological “ring” structure. The hybrid iiGA greatly outperforms the topological “ring” GA in terms of fitness and search efficiency for this given problem.

Epileptic Seizure Detection Using Genetically Programmed Artificial Features

Patient-specific epilepsy seizure detectors were designed based on the genetic programming artificial features algorithm, a general-purpose, methodic algorithm comprised by a genetic programming module and a k-nearest neighbor classifier to create synthetic features. Artificial features are an extension to conventional features, characterized by being computer-coded and may not have a known physical meaning. In this paper, artificial features are constructed from the reconstructed state-space trajectories of the intracranial EEG signals intended to reveal patterns indicative of epileptic seizure onset. The algorithm was evaluated in seven patients and validation experiments were carried out using 730.6 hr of EEG recordings. The results with the artificial features compare favorably with previous benchmark work that used a handcrafted feature. Among other results, 88 out of 92 seizures were detected yielding a low false negative rate of 4.35%